The overall goal of this proposal is to enhance our understanding of the importance of cyclooxygenase (Cox)-catalyzed reactions in the pathogenesis of breast cancer. Cox is potentially important because i) it converts chemicals to reactive electrophiles that are mutagenic and ii) it catalyzes the production of prostaglandins (PGs). PGs are believed to be important in the post-initiation phases of tumorigenesis. Particular attention will be directed toward Cox-2, a recently identified enzyme that can be induced by cytokines, growth factors and tumor promoters. We have shown that retinoids including N-(4-hydroxyphenyl) retinamide (4-HPR) downregulate the expression of the Cox-2 gene in mammary epithelial cells and inhibit the synthesis of PGs. The theoretical benefits of inhibiting Cox have been confirmed by studies showing that drugs that inhibit Cox, e.g. aspirin, protect against cancer. Our results establish an important mechanistic link between two major classes of chemopreventive agents: inhibitors of PG synthesis and retinoids. Moreover, our results are potentially important in explaining why retinoids are effective in preventing mammary cancer. The objectives of this proposal are the following: 1) Define the mechanism by which retinoids downregulate the expression of the Cox-2 gene in mammary epithelial cells. We will determine the importance of AP-1 transcription factors and estrogen receptor status in mediating this effect. 2) Enhance our understanding of the importance of Cox-catalyzed reactions in the pathogenesis of cancer. We will determine if increased levels of Cox-2 affect either the metabolism of xenobiotics to proximate carcinogens or cellular proliferation. 3) Determine whether increased expression of Cox-2 is responsible at least, in part, for the overproduction of PGE/2 in malignant breast tissue. Defining the basis for the increased production of PGs in malignant tissue could be an important step in developing new therapeutic approaches. It is important to stress that the results of these experiments are likely to be broadly applicable. We know, for example, that inhibitors of PG synthesis protect against colon cancer. Moreover, retinoids are being tested in multiple human chemoprevention trials.